TTLG|Thief|Bioshock|System Shock|Deus Ex|Mobile
Page 9 of 33 FirstFirst ... 4567891011121314192429 ... LastLast
Results 201 to 225 of 803

Thread: openDarkEngine

  1. #201
    Member
    Registered: Sep 2005
    Location: Prague
    Yes. OSM files can't be simply rewritten, maybe later on. They manage player-world interaction, objectives handling and so on. There is huge number of functions.

    They are not the engine itself, they only implement scripts. (Unfortunetaly written in C++ using COM)

    See Telliamed's great page for info if you like.
    http://thiefmissions.com/telliamed/

  2. #202
    Member
    Registered: Nov 2001
    Location: Sunny Finland
    Well, to some extent I agree with OrbWeaver's comment about the openness of such solution.

    The biggest problem with this project is that the end result should play pretty much exactly like classic Thief, otherwise what's the point? But to achieve that, you'd have to duplicate stuff like sound propagation and AI quite accurately. I don't see how that would be possible.

    But I'm not telling you to stop. On the contrary, even just being able to run around the familiar levels rendered by a more modern engine would be useful for something.

  3. #203
    Member
    Registered: Jun 2000
    Location: Finland, Earth
    Telliamed's been busy reverse-engineering the OSM scripts for a unified script library that AFAIK is intended to include all mission-specific and user-made scripts. I have no idea how far along it is or if he's planned to include the gen.osm scripts too, but it's on the way at least and would at least partly resolve any open-codebase issues.

  4. #204
    Member
    Registered: Nov 2002
    Location: Germany
    Quote Originally Posted by jermi
    The biggest problem with this project is that the end result should play pretty much exactly like classic Thief, otherwise what's the point? But to achieve that, you'd have to duplicate stuff like sound propagation and AI quite accurately. I don't see how that would be possible.
    Well, the same way that the original developers did it. By implementing it. But there is another big problem with this project. All that stuff has to be reverse engineered, which takes a lot of time. And the end result will be to have exactly the same Thief that we already have. Now another thing is the editor. What good is this engine if you can't create new levels with it? So you will have to replace Dromed as well. Either by reimplementing it in the same way, or by writing a new editor from scratch. This costs a LOT of time. I would expect that in the meantime TDM is long finished and people start to move to it over time. So the project would be interesting as a kind of proof of concept, but I don't really see that much benefit from it in the long run.

  5. #205
    Member
    Registered: Jun 2000
    Location: Finland, Earth
    From what I gather Volca intends the game engine to run the same missions as the original game, identically to how the original game ran them. Hence, missions created with Dromed would run also.
    The benefits of a reverse-engineered engine? Cross-platform support, bugfixes, removal of design limits, and compatible with a very large body of previously-developed fan content.

  6. #206
    Member
    Registered: May 2004
    With TDM it took us about 6 months just to design and implement an AI sound propagation system on par with T1/2. Reproducing the gameplay code for a game like Thief is not easy.

  7. #207
    Well I'm glad to see he's moved to Ogre 3D. It's probably the best OS engine out there. I've got much more faith in this project now.

  8. #208
    Member
    Registered: Nov 2002
    Location: Germany
    Quote Originally Posted by Vigil
    From what I gather Volca intends the game engine to run the same missions as the original game, identically to how the original game ran them. Hence, missions created with Dromed would run also.
    I know that they would run also. after all, that's the point of this project, right? My point was in the reverse direction. If we have an open source Thief engine, what good is it, if you still have only Dromed to use to create content for it?

    The benefits of a reverse-engineered engine? Cross-platform support, bugfixes, removal of design limits, and compatible with a very large body of previously-developed fan content.
    Unless the editor is also rewritten, you can hardly get around the design limits. After all, these limits also exist in the editor, so even if the engine could potentially handle more and different content, there is no way to create it, because Dromed will not support it at this point. And this is where my above statement comes from.

    Quote Originally Posted by Ishtvan
    With TDM it took us about 6 months just to design and implement an AI sound propagation system on par with T1/2. Reproducing the gameplay code for a game like Thief is not easy.
    Especially considering, that we were completely free in our implementation, becyause we don't have to interface with existing code, while this project would have to reproduce the original effects in almost the exact same way, in order to properly keep working with the existing stuff. So this means that it will take even longer, because you have to reverse engineer it, while we were free to choose how to implement it.

  9. #209
    Member
    Registered: Jun 2000
    Location: Finland, Earth
    Ah yes, those are good points sparhawk and I hadn't considered that Dromed's own engine implementation is entirely encapsulated in Dromed itself. Phoo.

  10. #210
    Member
    Registered: Nov 2002
    Location: Germany
    Still, I would really look forward to a native Linux version of Thief. I always hoped to use Wine for that, but somehow it doesn't work properly, and being able run Thief natively on Linux would be a big bonus in my book. But I think the project is pretty limited as long as there is no accompanying editor as well.

  11. #211
    Member
    Registered: Sep 2005
    Location: Prague
    It is true that without an editor, no new content could be made if dromed stops working, and because dromed is even bigger project than the engine (for example only the CSG and BSP calculations are not a simple thing, not talking about AI walk paths, etc...).

    This is why I consider this project to enable playing the old missions (e.g. only those in the original format).

    My effort is not to create a parallel project to the TDM, this would not be usefull much. If I would like to do something like this, then doing a tool used to export the original missions to a TDM compatible format would be a better idea.

    As I understand, an explanation of the purpose was needed.

    Sure the obstructions are huge, but I take it as a hobby project, and I like the chellenge.

  12. #212
    Member
    Registered: Jul 2001
    Location: cesspool
    Here is some screenshots.

  13. #213
    Member
    Registered: Nov 2002
    Location: Germany
    Looks good. Does this mean that loading level geometry works already?

  14. #214
    Member
    Registered: Jun 2000
    Location: Finland, Earth
    Why are the textures getting so washed-out?

  15. #215
    I don't visit this forum anymore, but lucky I get email notifications of threads I've replied to in the past.

    I'm really glad to see these screen shots

    What I'm really looking forward to is multiplayer co-op on Thief 1 and 2 fan missions.

    But this could happen two ways; With DarkFate, or with DarkMod if someone writes a map converter (doubtful, it'd be a huge task, our conversation systems etc would work nothing like DarkEngine's), and then someone implements multiplayer in DarkMod after we release it.
    Not sure which one.

    Yeah I noticed the textures look pretty washed out too. Like in http://darkfate.ru/files/media/audio...reenshot_2.jpg the green pixels on the floor really stand out. Did it look that bad in the original game? Can't remember.

  16. #216
    Clearinghouse
    Registered: Aug 2002
    Location: Siberia, Russia
    Why are the textures getting so washed-out?
    As I have understood from http://sourceforge.net/projects/opde :
    "The problem with textures is still not fixed, but i'm trying though".
    Who works in С++?

  17. #217
    Clearinghouse
    Registered: Aug 2002
    Location: Siberia, Russia


    If you want to look at the code checkout the CVS version.
    Last edited by clearing; 12th Sep 2006 at 05:46.

  18. #218
    Member
    Registered: Apr 2006

    Use Tenebrae : )

    Tenebrae is a hot rodded Quake 1 engine that does per-pixel lighting.

    http://tenebrae.sourceforge.net/

    I believe that Quake 3 has been open sourced. The curves would add a lot to levels. Radiant is open now too.

    I found this too.
    http://www.icculus.org/quake3/

    The Quake family games are very reliable, but there is a catch, quake compilers use constructive geometery only, not the air-brush situation found in Thief and Unreal family games.

    I am interesting in this. Thanks for the effort.

  19. #219
    New Member
    Registered: Jul 2005
    Volca, this is absolutely fantastic! If this reaches a point when the good old missions are playable we can have procedural textures and then the graphics will be much better than in TDS. Whilst the graphics in TDS are better than in T2 (and T1, of course) I was always bothered by how the shadows looked on the brick walls with deep cracks. A little bit O/T: To be honest I wish that TDS had never been released since the playability wasn't anywhere near that of T2 but having seen the graphics of TDS I could no longer enjoy T2 as much as before. But this indeed seems like something that will rescue Thief. I might be able to contribute something at some point - I have written two relatively simple 3D game engines as school exercises so I'll consider abandoning my dromed work in favour of this.

  20. #220
    Clearinghouse
    Registered: Aug 2002
    Location: Siberia, Russia
    I might be able to contribute something at some point
    nident, if you would like to read what is happening on the developer side, please register your e-mail on opde-devel mailing list.
    Here: https://lists.sourceforge.net/lists/listinfo/opde-devel

  21. #221
    Permanently Enlarged
    Registered: Dec 2005
    Location: I could care...but I won't.
    ive been learning a lot about C++ in the past 2 1/2 weeks since i got Sams Teach Yourself C++ In 21 Days, and figgered i might be able to help somewhat. im not a pro, but i do understand a lot now.

  22. #222
    Member
    Registered: Sep 2005
    Location: Prague
    Wow, I didn't realise a discussion is happening here. Luck I browsed in randomly.

    Thanks for the replies!

    The washed out textures seem to be caused by some implicit gamma correction the ogre/devil is doing when loading textures. The level display works perfectly otherwise, with two minor problems of missing underwater effect (dunno how to do this) and Displacement of the lightmaps (still few are a bit moved away from the place they should be).

    The good thing is people could start writing pixel shaders and other advanced stuff right now.

    I'm currently thinking up an universal make system (autotools are painful under windows), and hope to do 0.1.9 release soon. After this, the core services will get designed/made. Telliamed posted a very nice list of the core services that we'll needed to code. I hope to start with those soon.

    We'll welcome anyone who is willing to help!

  23. #223
    Member
    Registered: Feb 2002
    Location: Go is to the fountain.
    Ooh, pixel shaders! I want the lowdown on that one.

  24. #224
    Member
    Registered: Jan 2002
    Location: Germany
    Quote Originally Posted by Volca View Post
    I'm currently thinking up an universal make system (autotools are painful under windows)
    You might want to check out SCons or, alternatively, Crystal Space's build system, which provides automatic Visual Studio project file generation.

  25. #225
    Member
    Registered: Sep 2005
    Location: Prague
    @Erluk: Thanks for info! I'm currently succesfull with cmake (Linux/VC builds already), I'll be trying to make mingw/msys targets to work too. There also is Jam, which looks good. I hope to stick with the cmake as it looks promissing (if it will supply the needed platform independance).

    The face lifting will be a "fruits" part of the project. I was thinking about map export to some 3d editor, so people could edit the generated maps, adding details. Those would be an override for the internal map if found. Also dynamic shadows/global world dynamic illumination could be a nice thing. This is only a dream in this part of the project (But ogre3d enables one to do it quite easily).

Page 9 of 33 FirstFirst ... 4567891011121314192429 ... LastLast

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •